The IgG isotype of human antibodies consists of subtypes IgG1, IgG2, IgG3, and IgG4, each containing two antigen binding arms (Fabs) connected to a single Fc domain by the hinge region. IgG1, the predominant subclass represented in therapeutic monoclonal antibodies, are considered stable molecules with long half-life in circulation of 17.6 to 56.2 days (Salfeld, 2007 Nat Biotechnol 25:1369-72). However, IgG1 is susceptible to proteolysis in the hinge region by a number of physiologically-relevant proteases associated with invasive cancer (e.g. matrix metalloproteinases), inflammatory autoimmune diseases (e.g. MMP-3 and MMP-12 secretion in inflammatory bowel disease and rheumatoid arthritis) and pathological microorganisms. Cleavage above the disulfide bonds (core hinge) between the heavy chains liberates the monovalent Fab and bilateral cleavage below the disulfide bonds liberates a bivalent structure, the F(ab′)2 fragment. Several metalloproteinases and two bacterial enzymes, glutamyl endopeptidase V8 of Staphylococcus aureus (GluV8) and Immunoglobulin degrading enzyme of Streptococcus pyogenes (IdeS), act on IgG1 in the lower hinge (below the disulfide bonds (FIG. 1) and ultimately produce a F(ab′)2 and an Fc fragment (Ryan et al., 2008 Mol Immunol 45:1837-46). A single proteolytic cleavage in one of the heavy chain polypeptides of an IgG1 causes a loss of the IgG1 's ability to bind FcγRs and drive Fc-mediated effector functions (Brezski et al., 2009 Proc Natl Acad Sci USA 106:17864-9). Both single and multiple cleavages of therapeutic monoclonal antibodies may lead to species that bind target but have lost some or all efficacy.
The antibody effector functions mediated by the Fc domain are important in the overall therapeutic effect of the antibody (mAb) (Bibeau et al., 2009 J Clin Oncol 27:1122-9; Cartron et al., 2002 Blood 99:754-8; Musolino et al., 2008 J Clin Oncol 26:1789-96). The Fc domain of the antibody interacting with Fc gamma receptors (FcγR) expressed on immune cells, as well as Fc domain interactions with complement contribute to the action of several monoclonal antibodies (mAbs) directed against cell surface antigens. These interactions can lead to the elimination of the mAb targeted cell by antibody-dependent cellular cytotoxicity (ADCC), antibody-dependent cellular phagocytosis (ADCP), or complement-dependent cytotoxicity (CDC).
The Fc-dependent effector functions also include antibody-dependent cytokine release (ADCR). Some of the first Fc engineering efforts were aimed towards silencing Fc:FcγR interactions in order to abrogate unwanted cytokine release when the intent of the mAb was to suppress immune responses, as was the case with the anti-CD3 epsilon targeting muromomab-CD3 (reviewed in (Labrijn et al., 2008 Curr Opin Immunol 20:479-85).
Interactions between the Fc domain of antibodies and FcγRs can also influence the cell fate decisions of immune effector cells. Fc interactions with FcγRI on macrophages converted the macrophages from a pro-inflammatory phenotype into a regulatory phenotype (Sutterwala et al., 1998 J Exp Med 188:217-22) characterized by the secretion of the anti-inflammatory cytokine IL-10. Some of the anti-inflammatory properties attributed to IL-10 include both inhibition of antigen presentation and the expression of co-stimulatory molecules, blocking monocyte differentiation into dendritic cells (DC), inhibition of DC maturation, suppression of tumor cell killing by macrophages, and suppression of the release of pro-inflammatory cytokines such as IL-1, IL-6, IL-12, IFNγ, and TNF (reviewed in Mosser and Zhang, 2008 Immunol Rev 226:205-18). These effects diminish the ability of antigen presenting cells (APCs) to drive pro-inflammatory, Th1 immune responses. IL-10 can also sustain the ability of T regulatory cells to inhibit Th1 immune responses. It was recently suggested that monoclonal antibodies targeting tumor-associated antigens (e.g. the anti-EFGR mAb, cetuximab) can also induce macrophage IL-10 production, resulting in an anti-inflammatory, pro-tumor microenvironment and potential lack of efficacy of the anti-cancer therapeutics (Pander et al., 2011 Clin Cancer Res 17:5668-73).
Antibodies have the ability to induce the release of pro-inflammatory cytokines by interacting with FcγRs on PBMCs. IFNγ is a pro-inflammatory cytokine that can enhance macrophage tumor cell killing by increasing reactive nitrogen intermediates, augment cross-presentation, increase expression of co-stimulatory molecules and MHC I and MHC II, and drive Th1 cell differentiation. IFNγ can also directly inhibit the growth of tumor cells and virally infected cells (Ikeda et. Al., 2002 Cytokine Growth Factor Rev 13:95-109). Triggering FcγRIIIa on NK cells resulted in IFNγ production (Cassatella et al., 1989 J Exp Med 169:549-67). Opsonization of tumor cells with the human wild type IgG1 anti-HER2 mAb trastuzumab resulted in induction of IFNγ secretion from NK cells, and this secretion was influenced by the presence of IL-12 (Parihar et al., 2002 J Clin Invest 110:983-92). The ability of mAbs to induce NK cells to secrete IFNγ is considered beneficial for tumor-antigen targeting mAbs.
There is a need for protease resistant and effector-function retaining Fc antibody platforms that do not elicit IL-10 secretion by macrophages, (i.e. and thus do not promote macrophages to convert to anti-inflammatory regulator macrophages) for the treatment of cancer and infectious disease and other disease where the destruction of target cells or tissues is desired.